Oral Tradition, 15/1 (2000): 159-169

Beowulf as Epic

Joseph Harris

Cambridge, Massachusetts, has been home to two translators of the Kalevala in the twentieth century, and both furnished materials, however brief, for an understanding of how they might have compared the Finnish epic to the Anglo-Saxon Beowulf. (The present brief Canterbridgean contribution to the generic characterization of Beowulf, taking a hint from the heterogeneous genre make-up of the Kalevala, focuses chiefly on a complex narrative structure and its meaning.) The better known of the two translators was my distinguished predecessor, the English professor and philologist Francis Peabody Magoun (1895-1979). In his 1963 translation of the 1849 Kalevala, Magoun’s allusions, still strongly under the spell of the early successes of the oral-formulaic theory, are chiefly to shared reliance on formulaic diction, though he does also point out certain differences in the two epics’ application of this style (1963a:xvii, n. 1; xviii, n. 3). Magoun’s reference to “the Beowulf songs” (xviii, n. 3) in the plural, however, alludes to his belief that different folk variants on the life of the hero can still be discriminated in the epic. By the time of his 1969 translation of the Old Kalevala Magoun seems not too far from the current standard view (xiv):

. . . such a semi-connected, semi-cyclic work as the Kalevala or the received text of the Anglo-Saxon Béowulf, however oral in its genesis, cannot possibly be the product of oral composition. Unlettered singers create in response to an immediate, eagerly waiting audience; they do not compose cyclically, but episodically . . . . it is for that reason that I have been firmly persuaded that the Anglo-Saxon Béowulf, being cyclic in treating more than one episode in the titular hero’s life, must, like the Kalevala, be the work of a lettered person using the verses of traditional singers.

Magoun’s narratological terms, especially “cyclical,” are obscurely used, but in his articles it does emerge clearly that he thinks of the received text as containing three songs about Beowulf “soldered” together (his metaphor) by a “concatenator” (1958, 1960, 1963b). Though Magoun’s dissecting, neo- 160 JOSEPH HARRIS

Liedertheorie does bring out some real issues that unitarians ought to answer, this mechanical view of poetry separated him from his own students (e.g., Creed 1966) as well as from current scholars of Beowulf, such as John Foley, who think rather in subtler terms of the “oral-derived” product of a real poet, not a concatenator (e.g., Foley 1991). The second Cambridge translator was of a very different sort. A relic of turn-of-the-century Finnish immigration to western Massachusetts, Eino Friberg (1901-95) was a poet; educated to the M.A. level in philosophy at Harvard, he was a man of letters who never held an academic position and who, most remarkably, had been blind since childhood. When I knew him, briefly at the end of his life, he was far gone in a romantic spirituality, which, I believe, can already be heard in the terse and sometimes stark poetry of his translation. His highly oblique reference to Beowulf, which is unnamed but clearly discernible beneath the general wording, is framed as a contrast to the Kalevala (1988:12):

The difference is clear, although expressing it may be somewhat more tentative; other Germanic and Scandinavian folklore collections seem to represent reactions to the imposition of Christianity, and the consequent loss of a past way of life—a fond farewell look over the shoulder before taking a place in the historical and literate order of Christendom.

The Kalevala, on the other hand, appears at an historical juncture, “a time when history was ripe with destiny” (1988:18), that projects its influence into the future. Friberg would have deplored a word like “concatenation”; to his inner eye “the structural units of the runo-singers were already transformationally related before Lönnrot recorded them” (1988:20)—which may be a poet’s way of referring to a mystic teleological coherence in what I would call the network of discourses, that is, in tradition. The Kalevala is not the only epic that ends with an aeon marked by the coming of a new religion: for one could say the same of the Shahnama and of the less familiar Watunna, collected and compiled in twentieth- century Venezuela (de Civrieux 1980). Such a structure perhaps becomes available to the epic poet whenever history seems to terminate myth. But how exactly to characterize the glance over the shoulder in Beowulf? This has become one of the many controversial themes in the interpretation of the Old English epic.

BEOWULF AS EPIC 161

Function, Context and Genre

A representative contemporary effort is that of the Freudian James Earl in his recent book Thinking About “Beowulf”(1994). Earl’s first chapter argues that the Anglo-Saxon epic is “an act of cultural mourning” (47):

. . . in Beowulf Christianity appropriates the mythic eschatology of the Germans by historicizing it . . . eschatology is the poem’s very motivation. The world destroyed at the end of the poem is the heroic world, that pre- Christian world which in many respects had to be renounced by the Anglo-Saxons with the coming of Christianity. The poem is in large part a lament for those losses; and precisely a lament, for in the poem Anglo- Saxon culture seems to be mourning for its lost past. Mourning epitomizes the normal, healthy processes of relinquishing the past and coming to terms with its absence. To the reader of Beowulf it need hardly be argued that a culture can mourn for its past as an individual can. Mourning is in fact commonly experienced collectively.

The chapter, under the dramatic title “The Birth and Death of Civilization,” is couched in a rhetorically overheated style and is also overstated: Earl’s confident handling of what he calls the “eschatology of the Germans” is hardly justified, even by the chief witness, Völuspá. Nevertheless, most contemporary interpreters of Beowulf, including me, do now seem inclined to read it as culturally postheroic and retrospective. It does not explicitly announce the coming of a new age in some imagined equivalent of the advent of the King of Karelia in Lönnrot’s Runo 50, but Beowulf’s barrow on Hronesnes does seem to bury a past in anticipation of something new. This interpretative consensus, to the extent that it is a consensus, does not extend to the date, provenance, and political function of the poem. Since 1979 the respectable range of dating possibilities has been extended from, say, 680 well past 800 to around the date of the manuscript, that is about 1000 or a little later.1 Several prominent Beowulfians are now arguing for the tenth century and the court of Æthelstan or another successor of Alfred (e.g., Niles 1993). This date enables them to think in terms of correlating the poem with the founding of the English nation in the wars of Æthelstan and his successors and thus to explain the Danish subject matter politically. I am of an older school. I think the poem has the kind of cultural function we heard of from James Earl, but the political function would have been pre-

1 The “aeon” of 1979 is supplied by a paradigm-shifting conference; see Chase 1981.

162 JOSEPH HARRIS

national, perhaps that of the familar speculum regis. I belong among those who think there is a connection between the East Anglian royal family, the Wuffings, and the land of the Geatas, present-day Västergötland; and the East Anglian ship burial at Sutton Hoo, about 625, would seem to furnish a clue to the earliest milieu for the culture reflected by the poem.2 However, unlike Sam Newton, I am inclined to regard the poem as we know it as a product of the Mercian (Middle Anglian) court of Offa the Great (757-97) or of a successor such as Wiglaf (827-40).3 In the context of current Beowulf scholarship, therefore, I belong among those who see the poem as chronologically relatively early, but typologically late in terms of cultural and literary history—as opposed to, for example, Niles (1993), who favors a chronologically late poem that is nevertheless foundational in cultural terms. At least all students of the poem agree in calling it an epic. Or do they? That ambiguous term papers over several disagreements. First, I reserve the word “epic” for relatively long narrative poems of a certain dignity; this allows for a hermeneutically useful contrast with the lay. Thus Hildebrandslied or Atlakvi∂a, for example, as relatively short narrative poems, are lays and not epics. I am not claiming that the Continental and Scandinavian use of “epic” to mean “narrative” is unjustified by custom (in German, for example) and etymology, only that the predominant Anglo- American customary sense (a long narrative poem with certain other features, which, however, are less distinctive than length) allows for useful contrasts that are missing in the non-Anglo-American tradition. In my main effort to present Anglo-Saxon heroic poetry within a consistent literary- historical framework (Harris 1985), I largely follow Andreas Heusler’s lead in constructing an ascending scale of poems from the sophisticated lists of Widsith and Deor, through eulogy, to the classic heroic lay, a native oral genre that we know from the West Germanic fragments, from the “five old poems” of the Poetic Edda, and from such secondary reflections as the Finnsburg Episode in Beowulf (cf. Heusler 1941). The next level of literary magnitude and complexity is the long narrative, the epic. Heusler and his present-day followers, for example my teacher Theodore Andersson (especially in 1987), derive the inspiration for the Buchepos ultimately from Virgil, and I concur. But the poles of lay and epic are more immediately a useful hermeneutical tool, as Heusler’s notion of the origin of epic by “swelling” (Anschwellung) of a lay already makes clear (cf. Haug 1975). The original of the Finnsburg Fragment, for instance, may have been longer

2 On this topic see most recently Newton 1993.

3 See Harris 1985:264-66. BEOWULF AS EPIC 163

and more leisurely than Heusler’s reconstructed ideal of the oral lay provides for, but it is still enlightening to contrast it with a “book-epic” such as the Latin Waltharius and probably with the original of the English fragments of the Waldere epic. Beowulf, however, the crowning achievement of Old English heroic poetry, lies outside all these genres, including even the “epic”; it presents a further generic development, having, of course, its point of departure in epics such as Waldere. In comparison to the lay and to simple epic developed by amplification, Beowulf embodies an exceptionally complicated narrative; the essential story—the monster fights and Beowulf’s life— conveys far less conception of the poem than is the case for lay and simple epic. In texture and structure Beowulf distinguishes itself from Waldere and other extant Germanic narratives. Its special blend of Christian and secular- heroic ethics is also unique, though perhaps not different in kind from Waldere’s. As a typologically “late” work, Beowulf carries within it strong evidence of awareness of its literary predecessors and as a whole implies an attitude toward and perspective on those sources. Like The Canterbury Tales, The Waste Land, Finnegan’s Wake, and The Kalevala, great works that transcend inherited genres, it embodies a new literary form that is, nevertheless, emphatically oriented toward the literature of the past. In the absence of an established term, I have called this kind of work a summa litterarum (Harris 1982, 1985). Such summae are generically synthetic and punctuate or terminate a period, summarizing or summa-rizing the literary past and seeming either to generate no direct progeny or to devour their own by overshadowing them in the course of literary history. The Beowulfian summa represents the poet’s “reading” of antecedent literature, mostly oral literature; its unity is more organic and its “idea” less a critical problem than for The Canterbury Tales (cf. Howard 1976) or The Kalevala. Yet anthology-like, it contains at least the following genres: genealogical verse, a creation hymn, elegies, a lament, a heroic lay, a praise poem, historical poems, a flyting, heroic boasts, gnomic verse, a sermon or paternal advice, and perhaps less formal oral genres. In addition, a number of other genres are alluded to, just as Chaucer alludes to drama; but without paraphrase the generic terms (for example, spell) are difficult to interpret. As a whole, then, Beowulf represents for me a unique poet’s unique reception of the oral genres of the Germanic early middle ages.

164 JOSEPH HARRIS

Nested Narrative Parallels

From the beginning some of the most interesting work on Beowulf has been narratological. Connections between this formal or structural work and the kind of hermeneutics we sampled from James Earl are generally risky, but I would like to devote the remainder of my essay to taking just such a risk. Anyone who knows the text of Beowulf knows that its mode of storytelling “lacks steady advance,” to paraphrase the famous judgment of Frederick Klaeber (1950:lvii); there are all manner of digressions, predictions, and resumes, in addition to the repetition of whole story- structures. I would like to examine one type of repetition that has escaped structural description and that can, I venture, have a meaning coordinate with the sense of the whole. It is the structure of narrative repetition known as mise en abîme. The history of the term and its use in literary theory have been brilliantly chronicled by Lucien Dällenbach (1989). In medieval heraldry a shield that contains a replica of itself was said to contain it “en abîme”; closer to our own time and social level, the Quaker Oats box has (or used to have) on it a Quaker who holds up a box of Quaker Oats, which of course has on it a Quaker who holds up another, and so forth. The play within a play in Hamlet is a famous example in drama, and in narrative the mise en abîme has been described simply as the primary fabula containing, embedded within it, a secondary fabula that mirrors it (Bal 1985:143-48). Perhaps “nested narrative parallels” would be an adequate brief definition. In any case, the possibility of infinite regress seemingly offered by this structure suggests that in the mise en abîme we are led inward to the core of a story.4 While Beowulf would be a paradise for any student of repetition in narrative, it offers one passage that I believe immediately qualifies as a significant mise en abîme. Beowulf begins his final series of speeches with a reminiscence of his childhood in fosterage with his uncle King Hrethel, whose old age is soon blighted when one of his sons accidentally kills another (2425-43); in Chickering’s moving translation: “There was no way to pay for a death so wrong, / blinding the heart, yet still the prince / had lost his life, lay unavenged” (1977:2441-43). Hrethel suffered grief and frustration that no revenge could be taken against the killer “though he did

4 The ideas in this section are more extensively developed in Harris 2000. For an excellent further discussion, see Jefferson 1983. Beowulf is cited by line number from Klaeber 1950 or in the translation of Chickering 1977. BEOWULF AS EPIC 165

not love him” (2467b), with the result that the king soon passed from this world in death. Literally into the middle of this domestic tragedy—en abîme—is inserted an analogous tale attached by swa (“so”) at the beginning and end (2444-62a), in other words something like a Homeric simile. Here an old man who loses his son on the gallows (also an unavengeable death) grieves that he cannot help the lad. The boy’s room reminds him of his emptiness and the absence of former joys. Then the old man takes to his bed and sings a lament. We do not learn whether he dies like Hrethel or lives, saved by song, as Egill Skalla-Grímsson does in a similar incident in an Icelandic saga, but, like Egill, the Old Man’s intention is certainly to put life behind him (Harris 1994, 2000). No one is likely to disagree that a tale within a tale connected by the explicit marking of a simile can be called a “nested narrative parallel,” but I want to push the argument for a mise en abîme centered on the Old Man’s Lament to a further stage with the claim that we should recognize two more layers of this regress. I expect some resistance to this proposal. Within the simile comes a moment when the father thinks about and rejects an ancient maxim, a maxim that we know more directly from Egill Skalla-Grímsson’s elegiac poem for his dead sons, Sonatorrek, stanza seventeen of which observes: “This is also said that no one may get recompense for his son unless he himself begets yet another descendant who will be for the other a man born in place of his brother.”5 I have argued that both poets have in mind the same piece of gnomic wisdom, a survival from a primitive stage of the family in which there did exist a form of compensation, namely in rebirth, “Wiederverkörperung in der Sippe,” as Karl August Eckhardt put it (1937). The gnome is worded negatively, however, casting doubt on itself: “no one may get recompense for his son unless. . .” Egill is definitely quoting the saying sarcastically; though he does not announce his intention to die in the poem—it is the saga that elaborates that theme—it is clear that Egill is not about to await a replacement son. When the preconceptions of the Old English are teased out at length, as I did in the earlier paper (Harris 1994), it appears that the gamol ceorl’s (or Old Man’s) decision not to await a replacement son must mean that he knows some men would wait, that he is rejecting the ancient wisdom without quoting it. So the gamol ceorl’s actions are based on something very like the maxim quoted in Sonatorrek. Now I want to claim that the maxim carries with it the presupposition of an implicit story in potentia—a sort of freeze- dried narrative. It is not a series of events in the past tense (a narrative), but is framed as a condition and a consequence—the structure of a belief, a

5 For references and a detailed discussion, see Harris 1994. 166 JOSEPH HARRIS

superstition, or a legal scenario: “If a man lose his son, he may get no other recompense than to engender another to replace the first.” The transformational relationship between this story manqué and a biblical- sounding parable should be clear. This is the innermost repetition, the center of the simile—en abîme. Finally we turn to the outermost shell. The whole passage we have just examined is spoken by Beowulf, who is very old and approaching his death; the epic never reported a marriage for Beowulf, and it is certain that he has no son. After the dragon-fight and his mortal wounding, his nephew Wiglaf just manages to bring him back to consciousness; but the old king’s first words concern his lacking son: “Now I would want to give to my son / these war-garments, had it been granted / that I have a guardian born from my body / for this inheritance (Chickering 1977:2729-32a). “Guardian” here, yrfeweard, is the term used of the gamol ceorl, who did not care to await a replacement yrfeweard (2453a); the grieving father tends toward death, perhaps by implication toward a reunion with the dead son and other relatives, and Beowulf, with his missing son on his mind, “must go after” his departed kinsmen (2816). Hrethel leaves his possessions to his surviving sons (eaferum, 2470a), and Beowulf, after expressing his regret for not having a son for this function, finally gives his personal war-gear to Wiglaf (2809-12), but he does not adopt his nephew though he had admonished Wiglaf “to watch / the country’s needs” (2800b-01a). These motifs around the theme of sonlessness are complicated in a curious way by the virgin king’s treatment of his people as a son-substitute and of the dragon’s treasure as a patrimony. After regretting the lack of a son, Beowulf’s next topic is his people (2732b ff., 2794 ff.), and the treasures are left to them though they are not called his eaforas. Death, in either case, is the father’s part. When one begins to work on this text in terms of parallels and their meaning—and here I am thinking immediately of Wiglaf and the Last Survivor—it seems to offer no terminal point. But I end here with the suggestion that our quadrupled story about the death of a sonless father, literally nested parallel narratives, takes us en abîme, to the heart of a special kind of “epic” that projects cultural anxiety about the end of one era and problematic connections to the fictional future—that is, to the real audience’s more recent past.

Harvard University

BEOWULF AS EPIC 167

References

Andersson 1987 Theodore M. Andersson. A Preface to the Nibelungenlied. Stanford: Stanford University Press.

Bal 1985 Mieke Bal. Narratology: Introduction to the Theory of Narrative. Toronto: University Toronto Press.

Chase 1981 Colin Chase, ed. The Dating of Beowulf. Toronto: University of Toronto Press.

Chickering 1977 Howell D. Chickering ed. and trans. Beowulf: A Dual- Language Edition. Garden City: Anchor. de Civrieux 1980 Marc de Civrieux. Watunna: An Orinoco Creation Cycle. Ed. and trans. by David M. Guss. San Francisco: North Point. (Spanish original: Watunna: mitología makiritare. Caracas: Monte Avila, 1970).

Creed 1966 Robert P. Creed. “‘. . . wél-hwel¶ gecwæπ. . .’: The Singer as Architect.” Tennessee Studies in Literature, 11:131-43.

Dällenbach 1989 Lucien Dällenbach. The Mirror in the Text. Trans. by Jeremy Whiteley and Emma Huges. Chicago: University of Chicago Press. (French original: Le récit speculaire: Contribution à l’étude de la mise en abîme. Collection poétique. Paris: Seuil, 1977).

Earl 1994 James W. Earl. Thinking About “Beowulf.” Stanford: Stanford University Press.

Eckhardt 1937 Karl August Eckhardt. Irdische Unsterblichkeit: Germanischer Glaube an die Wiederverkörperung in der Sippe. Weimar: Böhlau.

Foley 1991 John Miles Foley. “Beowulf and the Old English Poetic Tradition.” In Immanent Art: From Structure to Meaning in Traditional Oral Epic. Bloomington: Indiana University Press. pp. 190-242.

Friberg 1988 Eino Friberg, trans. The Kalevala: Epic of the Finnish People. Ed. and intro. by George C. Schoolfield. 2nd ed. Helsinki: Otava.

168 JOSEPH HARRIS

Harris 1982 Joseph Harris. “Beowulf in Literary History.” Pacific Coast Philology, 17:16-23. Rpt. in Interpretations of Beowulf: A Critical Anthology. Ed. by R. D. Fulk. Bloomington: Indiana University Press, 1991. pp. 235-41.

Harris 1985 . “Die altenglische Heldendichtung.” In Europäisches Frühmittelalter. Ed. by Klaus von See. Neues Handbuch der Literaturwissenschaft, Bd. 6. Wiesbaden: AULA-Verlag. pp. 237-76.

Harris 1994 . “A Nativist Approach to Beowulf: The Case of Germanic Elegy.” In Companion to Old English Poetry. Ed. by Henk Aertsen and Rolf H. Bremmer. Amsterdam: VU University Press. pp. 45-62.

Harris 2000 . “‘Double scene’ and ‘mise en abîme’ in Beowulfian Narrative.” In Gudar på jorden. Festskrift till Lars Lönnroth. Ed. by Stina Hansson and Mats Malm. Stockholm / Stehag: Symposion. pp. 322-38.

Haug 1975 Walter Haug. “Andreas Heuslers Heldensagenmodell: Prämissen, Kritik und Gegenentwurf.” Zeitschrift für deutsches Altertum und deutsche Literatur, 104:273-92.

Heusler 1941 Andreas Heusler. Die altgermanische Dichtung. 2nd ed. rev. Potsdam: Athenaion. Rpt. Darmstadt: Gentner, 1957.

Howard 1976 Donald R. Howard. The Idea of the Canterbury Tales. Berkeley: University of California Press.

Jefferson 1983 Ann Jefferson. “Mise en abîme and the Prophetic in Narrative.” Style, 17:196-208.

Klaeber 1950 Fr[ederick] Klaeber. Beowulf and The Fight at Finnsburg. 3rd ed. with 1st and 2nd suppls. Boston: Heath.

Magoun 1958 Francis Peabody Magoun. “Béowulf A': A Folk-Variant.” Arv: Tidskrift för nordisk folkminnesforskning, 14:95-101.

Magoun 1960 . “Conceptions and Images Common to Anglo- Saxon Poetry and the Kalevala.” In Britannica: Festschrift für Hermann M. Flasdieck. Ed. by Wolfgang Iser and Hans Schabram. Heidelberg: Carl Winter. pp. 180-91.

Magoun 1963a , trans. The Kalevala or Poems of the Kaleva District. Compiled by Elias Lönnrot. Cambridge, MA: Harvard University Press. BEOWULF AS EPIC 169

Magoun 1963b . “Béowulf B: A Folk-Poem on Béowulf’s Death.” In Early English and Norse Studies Presented to Hugh Smith. Ed. by Arthur Brown and Peter Foote. London: Methuen. pp. 127-40.

Magoun 1969 , trans. The Old Kalevala and Certain Antecedents. Compiled by Elias Lönnrot. Cambridge, MA: Harvard University Press.

Newton 1993 Sam Newton. The Origins of Beowulf and the Pre-Viking Kingdom of East Anglia. Cambridge: Brewer.

Niles 1993 John D. Niles. “Locating Beowulf in Literary History.” Exemplaria, 5:79-109.